This application is for continuation of a training program in biostatistics with a specific focus on cancer research. This training program supports six students each year. The practice of biostatistics is changing, especially in the area of cancer research. Discoveries over the past two decades into the molecular, biochemical, and genetic components of cell growth and development have changed our understanding of what makes a normal cell become cancerous, and then later metastasize. Given numerous available treatments for many cancers, each with efficacy for a subset of patients, this information has the potential to be leveraged to personalize therapy so each patient can be given the treatment that is most likely to benefit them given their cancer's molecular signature. Early detection and prevention efforts have also made inroads in numerous cancers, and demonstrate continuing promise to reduce mortality and morbidity due to cancer. The development of new methods for surveying a tumor molecularly and genetically have led to enormous, complex data sets whose relevant biological information is difficult to extract, and the intricacy and exceeding sensi- tivty of many of these methods raise considerable challenges in terms of designing studies that obtain reproducible measurements and replicable results. Designing clinical trials that can incorporate such information and take it into account in patient treatment is challenging, and requires many quantitative issues be carefully handled. Multidisciplinary research teams are at the heart of modern approaches to fighting cancer, and biostatisticians and other quantitative scientists find themselves as an increasingly crucial part of such teams, given the fundamental quantitative nature of many of the modern biomedical research challenges. Such teamwork requires that the statistician be conversant with molecular biologists, pathologists, and, of course, clinical oncologists if the group of co-investigators is to succeed. Effective biostatisticl collaboration thus requires broad training in statistics, probability, computational methods, as well as cancer biology and medical ethics. The Department of Statistics, Rice University, and the departments in the Division of Quantitative Sciences at the University of Texas M. D. Anderson Cancer Center have joined forces with their collaborators in the clinical and basic sciences to developed a unique training program that combines their respective strengths to train biostatisticians in cancer research. The goal of this Training Program in Biostatistics is to prepare a new generation of biostatisticians who will work side-by-side with biomedical investigators in modern cancer research. Our program aims to provide trainees with: - Rigorous training in statistics and probability; - Practical experience in basic and clinical cancer research; - Training in biological aspects of cancer and medical research ethics. Students in this program follow a standard course of study expected of Ph.D. students in the Department of Statistics at Rice, with additional coursework in biostatistics, biology, and ethics, as well as special seminars and workshops at both institutions. With faculty expertise in Bayesian methods, decision theory, cancer clinical trials, cancer screening, survival analysis, statistical genetics, genomics, bioinformatics, and statistical computing, trainees will receive a broad background in biostatistics necessary for modem cancer research. Summer internships and laboratory rotations provide practical experience in the student's training.